Creative software developer Adobe Inc. is rolling out three new generative artificial intelligence tools for Premiere Pro, the company’s video editing software, designed to assist video editors with their creative workflows that will be coming sometime later this year.

The new tools will be joining Adobe’s already existing AI models in the company’s Firefly family of models, which power creative design expression using generative AI within several of its other products across Adobe Creative Cloud and Adobe Express. For example, Firefly can be used in Photoshop to create art using text-to-image to fill in or change an image according to what the user typed.

The new video tools add generative extend that lengthens clips, object addition and removal within video and text to video generation.

“Adobe is reimagining every step of video creation and production workflow to give creators new power and flexibility to realize their vision,” said Ashley Still, senior vice president of the creative product group at Adobe.

Generative extend allows editors to seamlessly extend video clip content using the generative AI model so it can be transitioned easily into another clip. This capability helps solve a problem that editors routinely run into daily when attempting to tie together different clips and need to edit them together when one clip is cut short, thus creating a jarring experience for viewers.

Object addition and removal allow editors to easily select and track objects in the video scene and then replace them with whatever they want using a text prompt. This makes it simple to remove an unwanted item or even change an actor’s wardrobe without advanced editing, the same tool can be used to alter or add set dressings, such as paint or photorealistic flowers.

Text-to-video can generate entirely new footage directly within the Premiere Pro app using text prompts or uploaded reference images. These short clips can be used to create storyboards, B-roll or to augment action footage.

“By bringing generative AI innovations deep into core Premiere Pro workflows, we are solving real pain points that video editors experience every day while giving them more space to focus on their craft,” added Still.

Some of Firefly’s generative AI capabilities are already integrated in Premiere Pro, such as being able to change color schemes of existing footage, generating royalty-free custom sounds and music, and the ability to automatically producing custom font overlays of logos and title cards.  The new features will add onto these existing capabilities that will expand into the ability to generate new videos from scratch.

The new Firefly generative AI video model represents Adobe’s response to video-producing AI models in the industry such as OpenAI’s Sora and Runway AI Inc.’s models, which can create short video clips from user prompts. The company also previewed a vision for integrating third-party models from OpenAI, Runway and Pika Labs into its video editing software as part of its plugin ecosystem to provide creators more options for AI video creation opportunities.

Image: Adobe